Supervised spatial classification of multispectral LiDAR data in urban areas.
Ontology highlight
ABSTRACT:
SUBMITTER: Huo LZ
PROVIDER: S-EPMC6200265 | biostudies-literature | 2018
REPOSITORIES: biostudies-literature
Publications
Supervised spatial classification of multispectral LiDAR data in urban areas.
Huo Lian-Zhi LZ Silva Carlos Alberto CA Klauberg Carine C Mohan Midhun M Zhao Li-Jun LJ Tang Ping P Hudak Andrew Thomas AT
PloS one 20181024 10
Multispectral LiDAR (light detection and ranging) data have been initially used for land cover classification. However, there are still high classification uncertainties, especially in urban areas, where objects are often mixed and confounded. This study investigated the efficiency of combining advanced statistical methods and LiDAR metrics derived from multispectral LiDAR data for improving land cover classification accuracy in urban areas. The study area is located in Oshawa, Ontario, Canada, ...[more]
Similar Datasets
Project description:Geographic information systems (GIS) provide accurate maps of terrain, roads, waterways, and building footprints and heights. Aircraft, particularly small unmanned aircraft systems (UAS), can exploit this and additional information such as building roof structure to improve navigation accuracy and safely perform contingency landings particularly in urban regions. However, building roof structure is not fully provided in maps. This paper proposes a method to automatically label building roof shape from publicly available GIS data. Satellite imagery and airborne LiDAR data are processed and manually labeled to create a diverse annotated roof image dataset for small to large urban cities. Multiple convolutional neural network (CNN) architectures are trained and tested, with the best performing networks providing a condensed feature set for support vector machine and decision tree classifiers. Satellite image and LiDAR data fusion is shown to provide greater classification accuracy than using either data type alone. Model confidence thresholds are adjusted leading to significant increases in models precision. Networks trained from roof data in Witten, Germany and Manhattan (New York City) are evaluated on independent data from these cities and Ann Arbor, Michigan.
| S-EPMC6264004 | biostudies-literature
Project description:Urban areas feature complex and heterogeneous land covers which create challenging issues for tree species classification. The increased availability of high spatial resolution multispectral satellite imagery and LiDAR datasets combined with the recent evolution of deep learning within remote sensing for object detection and scene classification, provide promising opportunities to map individual tree species with greater accuracy and resolution. However, there are knowledge gaps that are related to the contribution of Worldview-3 SWIR bands, very high resolution PAN band and LiDAR data in detailed tree species mapping. Additionally, contemporary deep learning methods are hampered by lack of training samples and difficulties of preparing training data. The objective of this study was to examine the potential of a novel deep learning method, Dense Convolutional Network (DenseNet), to identify dominant individual tree species in a complex urban environment within a fused image of WorldView-2 VNIR, Worldview-3 SWIR and LiDAR datasets. DenseNet results were compared against two popular machine classifiers in remote sensing image analysis, Random Forest (RF) and Support Vector Machine (SVM). Our results demonstrated that: (1) utilizing a data fusion approach beginning with VNIR and adding SWIR, LiDAR, and panchromatic (PAN) bands increased the overall accuracy of the DenseNet classifier from 75.9% to 76.8%, 81.1% and 82.6%, respectively. (2) DenseNet significantly outperformed RF and SVM for the classification of eight dominant tree species with an overall accuracy of 82.6%, compared to 51.8% and 52% for SVM and RF classifiers, respectively. (3) DenseNet maintained superior performance over RF and SVM classifiers under restricted training sample quantities which is a major limiting factor for deep learning techniques. Overall, the study reveals that DenseNet is more effective for urban tree species classification as it outperforms the popular RF and SVM techniques when working with highly complex image scenes regardless of training sample size.
| S-EPMC6471063 | biostudies-literature
Project description:PurposeUrban areas are usually found to have higher rates of self-harm, with deprivation the strongest predictor at area-level. We use a disease mapping approach to examine how self-harm is patterned within an urban area and its associations with deprivation, urbanness and ethnicity.MethodsData from clinical records on individuals admitted for self-harm for 725 small areas in South East London were included. Bayesian hierarchical models explored the spatio-temporal patterns of self-harm admission rates and potential associations with proximity to city centre, population density, percentage greenspace and non-white ethnic-minority populations. All models were adjusted for area-level deprivation, social fragmentation and hospital of admission.ResultsThere were 8327 first admissions for self-harm during the study period. Self-harm admission rates varied fourfold across the study area, with lower rates close to the city centre [adjusted standardised admission ratio, closest versus furthest quartile 0.71(95% CrI 0.54-0.96)]. Deprivation was associated with self-harm but partially masked rather than explained the spatial pattern, which strengthened after adjustment. After adjustment for deprivation, hospital of admission and social fragmentation, greenspace, population density and ethnicity were not associated with self-harm rates.ConclusionProximity to the city centre was associated with lower rates of self-harm, but the usual operationalisations of urbanness, population density and greenspace, were not. Deprivation did not explain the spatial patterning, nor did ethnicity. While nationally self-harm rates are higher in urban and deprived areas, this cannot be extrapolated to mean that within cities the inner-city is the highest risk area nor that risk will be principally patterned according to deprivation.
| S-EPMC6336929 | biostudies-literature
Project description:Accurately measuring the variability of spectroscopic signatures of hydration (H2O + OH) on the illuminated lunar surface at 3 μm as a function of latitude, lunar time of day, and composition is crucial to determining the generation and destruction mechanisms of OH species and understanding the global water cycle. A prime complication in analysis of the spectroscopic feature is the accurate removal of thermal emission, which can modify or even eliminate the hydration feature depending on the data processing methods used and assumptions made. An orbital multispectral lidar, with laser illumination at key diagnostic wavelengths, would provide uniform, zero-phase geometry, complete latitude and time of day coverage from a circular polar orbit, and is agnostic to the thermal state of the surface. We have performed measurement simulations of a four-wavelength multispectral lidar using spectral mixtures of hydrated mid-ocean-ridge basalt (MORB) glasses and lunar regolith endmembers to assess the lidar performance in measuring hydration signatures on the lunar surface. Our results show a feasible system with wavelengths at 1.5 μm, 2.65 μm, 2.8 μm, and 3.1 μm can measure lunar hydration with a precision of 52 ppm (1σ) or better. These results, combined with the uniform measurement capabilities of multispectral lidar make it a valuable spectroscopic technique for elucidating mechanisms of OH/H2O generation, migration, and destruction.
| S-EPMC9400864 | biostudies-literature
Project description:An international data science challenge, called National Ecological Observatory Network-National Institute of Standards and Technology data science evaluation, was set up in autumn 2017 with the goal to improve the use of remote sensing data in ecological applications. The competition was divided into three tasks: (1) individual tree crown (ITC) delineation, for identifying the location and size of individual trees; (2) alignment between field surveyed trees and ITCs delineated on remote sensing data; and (3) tree species classification. In this paper, the methods and results of team Fondazione Edmund Mach (FEM) are presented. The ITC delineation (Task 1 of the challenge) was done using a region growing method applied to a near-infrared band of the hyperspectral images. The optimization of the parameters of the delineation algorithm was done in a supervised way on the basis of the Jaccard score using the training set provided by the organizers. The alignment (Task 2) between the delineated ITCs and the field surveyed trees was done using the Euclidean distance among the position, the height, and the crown radius of the ITCs and the field surveyed trees. The classification (Task 3) was performed using a support vector machine classifier applied to a selection of the hyperspectral bands and the canopy height model. The selection of the bands was done using the sequential forward floating selection method and the Jeffries Matusita distance. The results of the three tasks were very promising: team FEM ranked first in the data science competition in Task 1 and 2, and second in Task 3. The Jaccard score of the delineated crowns was 0.3402, and the results showed that the proposed approach delineated both small and large crowns. The alignment was correctly done for all the test samples. The classification results were good (overall accuracy of 88.1%, kappa accuracy of 75.7%, and mean class accuracy of 61.5%), although the accuracy was biased toward the most represented species.
| S-EPMC6330952 | biostudies-literature
Project description:In this paper, we propose the quantum semi-supervised generative adversarial network (qSGAN). The system is composed of a quantum generator and a classical discriminator/classifier (D/C). The goal is to train both the generator and the D/C, so that the latter may get a high classification accuracy for a given dataset. Hence the qSGAN needs neither any data loading nor to generate a pure quantum state, implying that qSGAN is much easier to implement than many existing quantum algorithms. Also the generator can serve as a stronger adversary than a classical one thanks to its rich expressibility, and it is expected to be robust against noise. These advantages are demonstrated in a numerical simulation.
| S-EPMC8490428 | biostudies-literature
Project description:Airborne remote sensing offers unprecedented opportunities to efficiently monitor vegetation, but methods to delineate and classify individual plant species using the collected data are still actively being developed and improved. The Integrating Data science with Trees and Remote Sensing (IDTReeS) plant identification competition openly invited scientists to create and compare individual tree mapping methods. Participants were tasked with training taxon identification algorithms based on two sites, to then transfer their methods to a third unseen site, using field-based plant observations in combination with airborne remote sensing image data products from the National Ecological Observatory Network (NEON). These data were captured by a high resolution digital camera sensitive to red, green, blue (RGB) light, hyperspectral imaging spectrometer spanning the visible to shortwave infrared wavelengths, and lidar systems to capture the spectral and structural properties of vegetation. As participants in the IDTReeS competition, we developed a two-stage deep learning approach to integrate NEON remote sensing data from all three sensors and classify individual plant species and genera. The first stage was a convolutional neural network that generates taxon probabilities from RGB images, and the second stage was a fusion neural network that "learns" how to combine these probabilities with hyperspectral and lidar data. Our two-stage approach leverages the ability of neural networks to flexibly and automatically extract descriptive features from complex image data with high dimensionality. Our method achieved an overall classification accuracy of 0.51 based on the training set, and 0.32 based on the test set which contained data from an unseen site with unknown taxa classes. Although transferability of classification algorithms to unseen sites with unknown species and genus classes proved to be a challenging task, developing methods with openly available NEON data that will be collected in a standardized format for 30 years allows for continual improvements and major gains for members of the computational ecology community. We outline promising directions related to data preparation and processing techniques for further investigation, and provide our code to contribute to open reproducible science efforts.
| S-EPMC8325917 | biostudies-literature
Project description:BackgroundThe World Health Organization estimates that the global number of dengue infections range between 80-100 million per year, with some studies estimating approximately three times higher numbers. Furthermore, the geographic range of dengue virus transmission is extending with the disease now occurring more frequently in areas such as southern Europe. Ae. aegypti, one of the most prominent dengue vectors, is endemic to the far north-east of Australia and the city of Cairns frequently experiences dengue outbreaks which sometimes lead to large epidemics.MethodA spatially-explicit, individual-based mathematical model that accounts for the spread of dengue infection as a result of human movement and mosquito dispersion is presented. The model closely couples the four key sub-models necessary for representing the overall dynamics of the physical system, namely those describing mosquito population dynamics, human movement, virus transmission and vector control. Important features are the use of high quality outbreak data and mosquito trapping data for calibration and validation and a strategy to derive local mosquito abundance based on vegetation coverage and census data.ResultsThe model has been calibrated using detailed 2003 dengue outbreak data from Cairns, together with census and mosquito trapping data, and is shown to realistically reproduce a further dengue outbreak. The simulation results replicating the 2008/2009 Cairns epidemic support several hypotheses (formulated previously) aimed at explaining the large-scale epidemic which occurred in 2008/2009; specifically, while warmer weather and increased human movement had only a small effect on the spread of the virus, a shorter virus strain-specific extrinsic incubation time can explain the observed explosive outbreak of 2008/2009.ConclusionThe proof-of-concept simulation model described in this study has potential as a tool for understanding factors contributing to dengue spread as well as planning and optimizing dengue control, including reducing the Ae. aegypti vector population and for estimating the effectiveness and cost-effectiveness of future vaccination programmes. This model could also be applied to other vector borne viral diseases such as chikungunya, also spread by Ae. aegypti and, by re-parameterisation of the vector sub-model, to dengue and chikungunya viruses spread by Aedes albopictus.
| S-EPMC4152583 | biostudies-literature
Project description:Strawberry is one of the most popular fruits in the market. To meet the demanding consumer and market quality standards, there is a strong need for an on-site, accurate and reliable grading system during the whole harvesting process. In this work, a total of 923 strawberry fruit were measured directly on-plant at different ripening stages by means of bioimpedance data, collected at frequencies between 20 Hz and 300 kHz. The fruit batch was then splitted in 2 classes (i.e. ripe and unripe) based on surface color data. Starting from these data, six of the most commonly used supervised machine learning classification techniques, i.e. Logistic Regression (LR), Binary Decision Trees (DT), Naive Bayes Classifiers (NBC), K-Nearest Neighbors (KNN), Support Vector Machine (SVM) and Multi-Layer Perceptron Networks (MLP), were employed, optimized, tested and compared in view of their performance in predicting the strawberry fruit ripening stage. Such models were trained to develop a complete feature selection and optimization pipeline, not yet available for bioimpedance data analysis of fruit. The classification results highlighted that, among all the tested methods, MLP networks had the best performances on the test set, with 0.72, 0.82 and 0.73 for the F[Formula: see text], F[Formula: see text] and F[Formula: see text]-score, respectively, and improved the training results, showing good generalization capability, adapting well to new, previously unseen data. Consequently, the MLP models, trained with bioimpedance data, are a promising alternative for real-time estimation of strawberry ripeness directly on-field, which could be a potential application technique for evaluating the harvesting time management for farmers and producers.
| S-EPMC8160339 | biostudies-literature
Project description:BackgroundIn this study, a cross-sectional survey was used to understand and analyze the health status and health needs of the elderly in the community. The cluster analysis method was used to explore the relationship between health needs items and investigate the commonness among health demand items, to provide a reference for the development of health management of the elderly with chronic diseases.MethodsWe used convenience sampling to recruit the participants (aged 60 and above) from four urban community centers in Jinzhou City, Liaoning Province, China, in this study. This study uses the Medical Outcomes Study(MOS)36-Item Short-Form Health Survey. The self-designed questionnaire includes sociodemographic characteristics, chronic diseases, physical examination conditions, illness in the past two weeks, and a health needs questionnaire. SPSS 18.0 was used for data entry and analysis. Data analysis methods included descriptive statistical analysis, t-test, one-way analysis, cluster analysis, and linear multiple regression analysis.ResultsThe rate of health needs among the elderly in the community for various health services is 1.3-69.7%, of which the top three are: regular physical examination (69.7%), day Care Center (67.7%), the establishment of a Medical Alert Systems (66.1%). The health needs of the elderly in the community are divided into three categories: basic needs (24 items), health education (13 items), and first aid (2 items). The regression analysis found that the influencing factors of health status were age, revenue and expenditure, medical expenses, health education, basic needs, and first aid.ConclusionsThe community should strengthen the management of chronic diseases of the elderly and the publicity and education of related knowledge, and provide complementary health care services according to the health needs of the elderly, improve the health of the elderly, and improve the quality of life of the elderly.
| S-EPMC10563358 | biostudies-literature